Large scale refolding and purification of the catalytic domain of human BACE-2 produced in E. coli.

Evidence for a key role of beta-amyloid (Abeta) in Alzheimer's disease has led to considerable interest in potential therapeutic strategies targeting enzymes involved in processing the amyloid precursor protein (APP). Beta-site APP Cleaving Enzyme (BACE or beta-secretase) is a membrane bound aspartyl protease that has been shown to be directly involved in Abeta production and, therefore, is at the forefront of therapeutic targets in the treatment of Alzheimer's disease. BACE-2, an enzyme closely related to BACE, regulates Abeta production in a manner antagonistic to BACE, suggesting that non-selective inhibition of BACE-2 by BACE inhibitors might impair the lowering of Abeta. The design of BACE inhibitors that do not inhibit BACE-2 would be enhanced by structural and kinetic studies, efforts that typically demand considerable amounts of both enzymes. A BACE-2 construct containing 19 residues of the BACE prosegment followed by the BACE-2 catalytic domain sequence, Asp36-Trp447, was produced in E. coli inclusion bodies (IB) at 110-140 mg/L cell culture. Exploration of a variety of refolding conditions resulted in an efficient method for refolding the resulting pro-BACE-2 construct, and this protein undergoes facile autocatalytic cleavage, optimal at pH 4, at the Leu40- downward arrow-Ala41 bond. Refolded BACE-2 was purified by anion exchange, molecular sieve, and affinity chromatographies, yielding 105 mg of homogeneous enzyme (kcat/ Km = 1.2 x 10(4) x M(-1) x sec(-1)) from 8 liters of E. coli cell culture.

High yield expression of human BACE constructs in Eschericia coli for refolding, purification, and high resolution diffracting crystal forms.

BACE (beta-site APP cleaving enzyme) or beta-secretase, the enzyme responsible for processing APP to give the N-terminal portion of the Abeta peptide, is a membrane bound aspartyl protease consisting of an ectodomain catalytic unit, a C-terminal transmembrane segment and a cytoplasmic domain. Three BACE constructs, pET11a-BACE, pQE80L-BACE, and pQE70-BACE were designed to terminate at a position just before the transmembrane domain (Ser(432)) and are described schematically below. (1) pET11a-T7.Tag-G-S-M-(A-8GV......QTDES(432)), (2) pQE80L-Met-R-G-S-(His)(6)-G-S-I-E-T-D-(T(1)QH...QTDES(432)), and (3) pQE70-Met-BACE (R(36)GSFVEMG....PQTDES(432) (His) (6)) Each construct was over-expressed in Escherichia coli as inclusion bodies. The inclusion body proteins were solubilized in urea and refolded by dilution in water to yield active enzyme. Maximal activity for pET11a-BACE and pQE80L-BACE was usually reached at day 3 to 4, while construct pQE70-BACE required about 21 days. Active BACE was purified to homogeneity by anion-exchange chromatography and affinity chromatography over a column of immobilized peptide inhibitor. The process, easily scalable to 60 liters of cell culture, yielded in excess of 400 mg of active enzyme for crystallographic analysis. Highly purified pET11a-BACE and pQE70-BACE formed complexes with various inhibitors, the latter protein giving crystals diffracting up to 1.45 A resolution. In addition, a crystal form that does not require the presence of an inhibitor has been obtained for pQE70-BACE. This ligand-free crystal form has proven useful for the preparation of BACE-inhibitor complexes in soaking experiments.

Use of the multiple sclerosis functional composite as an outcome measure in a phase 3 clinical trial.

BACKGROUND: The Multiple Sclerosis Functional Composite (MSFC) is a multidimensional clinical outcome measure that includes quantitative tests of leg function/ambulation (Timed 25-Foot Walk), arm function (9-Hole Peg Test), and cognitive function (Paced Auditory Serial Addition Test). The MSFC is the primary outcome measure in the ongoing multinational phase 3 trial of interferon beta-1a (Avonex) in patients with secondary progressive MS. OBJECTIVE: To assess the practice effects, reliability, and validity of the MSFC clinical outcome measure. DESIGN: Examining technicians underwent formal training using standardized materials. The MSFC was performed according to a standardized protocol. The 436 patients enrolled in the International Multiple Sclerosis Secondary Progressive Avonex Controlled Trial underwent 3 prebaseline MSFC testing sessions before randomization. RESULTS: Practice effects were evident initially for the MSFC but stabilized by the fourth administration. The Paced Auditory Serial Addition Test demonstrated the most prominent practice effects. The reliability of the MSFC was excellent, with an intraclass correlation coefficient for session 3 (final prebaseline session) vs session 4 (baseline) of 0.90. The MSFC at baseline correlated moderately strongly with the Kurtzke Expanded Disability Status Scale. Among the MSFC components, the Timed 25-Foot Walk correlated most closely. Correlations among the 3 MSFC components were weak, suggesting they assess distinct aspects of neurologic function in patients with MS. CONCLUSIONS: The MSFC demonstrated excellent intrarater reliability in this multinational phase 3 trial. Three prebaseline testing sessions were sufficient to compensate for practice effects. The pattern of correlations among the MSFC, its components, and the Kurtzke Expanded Disability Status Scale supported the validity of the MSFC.

Recombinant human cytomegalovirus protease with a C-terminal (His)6 extension: purification, autocatalytic release of the mature enzyme, and biochemical characterization.

Human cytomegalovirus protease (CMV PR) is a target for the development of antiviral therapeutics. To obtain large amounts of native protease, a 268-amino-acid polypeptide with a hexahistidinyl tag at the C terminus was expressed in Escherichia coli. The first 262 amino acids of the recombinant protein were identical to the amino acid sequence of native CMV PR, except for mutations introduced at the internal cleavage site to eliminate autoproteolysis at that site. The hexahistidinyl tag was placed downstream of amino acid 262 of the native CMV PR sequence. In this design, the Ala-Ser bond at amino acids 256-257 constitutes a site naturally cleaved by the protease during capsid maturation. The 268-amino-acid polypeptide with the (His)6 tag was expressed at high levels in E. coli as inclusion bodies. After solubilization of the inclusion bodies, the protease was purified to homogeneity by a single step using Ni2+ affinity chromatography. The protease was refolded to an active enzyme using dialysis which leads to effective autocleavage of the Ala-Ser bond at amino acids 256-257 to remove 12 amino acids including the (His)6 tag from the C terminus of the protein. This strategy yielded large amounts of highly purified CMV PR with the native N terminus and C terminus. Approximately 40 mg of purified CMV PR was obtained per liter of cell culture using this strategy. The enzymatic activity of CMV PR purified from inclusion bodies and refolded to an active enzyme was similar to the enzymatic activity of CMV PR expressed as a soluble protein in E. coli. In addition, the refolded CMV PR could be crystallized for X-ray diffraction.

The HIV-1 protease as enzyme and substrate: mutagenesis of autolysis sites and generation of a stable mutant with retained kinetic properties.

Site-directed mutagenesis of autolysis sites in the human immunodeficiency virus type 1 (HIV-1) protease was applied in an analysis of enzyme specificity; the protease served, therefore, as both enzyme and substrate in this study. Inspection of natural substrates of all retroviral proteases revealed the absence of beta-branched amino acids at the P1 site and of Lys anywhere from P2 through P2'. Accordingly, several mutants of the HIV-1 protease were engineered in which these excluded amino acids were substituted at their respective P positions at the three major sites of autolysis in the wild-type protease (Leu5-Trp6, Leu33-Glu34, and Leu63-Ile64), and the mutant enzymes were evaluated in terms of their resistance to autodegradation. All of the mutant HIV-1 proteases, expressed as inclusion bodies in Escherichia coli, were enzymatically active after refolding, and all showed greatly diminished rates of cleavage at the altered autolysis sites. Some, however, were not viable enzymatically because of poor physical characteristics. This was the case for mutants having Lys replacements of Glu residues at P2' and for another in which all three P1 leucines were replaced by Ile. However, one of the mutant proteases, Q7K/L33I/L63I, was highly resistant to autolysis, while retaining the physical properties, specificity, and susceptibility to inhibition of the wild-type enzyme. Q7K/L33I/L63I should find useful application as a stable surrogate of the HIV-1 protease. Overall, our results can be interpreted relative to a model in which the active HIV-1 protease dimer is in equilibrium with monomeric, disordered species which serve as the substrates for autolysis.

Large scale purification and refolding of HIV-1 protease from Escherichia coli inclusion bodies.

The protease encoded by the human immunodeficiency virus type 1 (HIV-1) was engineered in Escherichia coli as a construct in which the natural 99-residue polypeptide was preceded by an NH2-terminal methionine initiator. Inclusion bodies harboring the recombinant HIV-1 protease were dissolved in 50% acetic acid and the solution was subjected to gel filtration on a column of Sephadex G-75. The protein, eluted in the second of two peaks, migrated in SDS-PAGE as a single sharp band of M(r) approximately 10,000. The purified HIV-1 protease was refolded into an active enzyme by diluting a solution of the protein in 50% acetic acid with 25 volumes of buffer at pH 5.5. This method of purification, which has also been applied to the purification of HIV-2 protease, provides a single-step procedure to produce 100 mg quantities of fully active enzyme.